Spatial Optimization of the Synoptic Sampling Network in the Great Smoky Mountains National Park

A water quality monitoring network consisting of 83 sampling sites in the Great Smoky Mountains National Park (GRSM) was optimized using a combination of multivariate statistical techniques and simulated annealing (SA). Principal components analysis (PCA), cluster analysis (CA), and discriminant analysis (DA) were used to group similar sampling sites in terms of water quality, watershed vegetation, watershed geology, and watershed geomorphology information. An objective function, incorporating the statistical results and the network costs, was formulated where the objective function was maximized for the benefits of the entire monitoring network using SA.